has expanded far beyond these limits, especially since the 1970s.
As already indicated, cattle have been the major asset – economically, politically and symbolically – since precolonial times and before the diamond era. Although crucial for postcolonial state formation before the state treasury became sourced by revenues from diamond mining, livestock production – like diamond mining – is highly labour intensive. While cattle have, in addition, always been highly unequally distributed, a very limited part of the population has made a living entirely from livestock production. This feature has become ever more pronounced in postcolonial times (Gulbrandsen 1996a: Ch. 10).
Since the late nineteenth century, vast numbers of men have ensured the survival of vast parts of the families in the country through circular labour migration, predominantly to the South African mines (Schapera 1947a; Gulbrandsen 1996a). This employment pattern was drastically curtailed in the beginning of the 1980s with a substantial reduction in recruitment of foreign labourers. About the same time, however, the development of urban areas in Botswana accelerated, involving a building boom that recruited many of those who had previously gone to South Africa. Especially the capital of Gaborone, which was established from scratch at independence, entered a process of momentous growth, propelled by the substantial enlargement of governmental institutions and the rapid expansion of the private sector of the economy from the late 1980s onwards. Nevertheless, unemployment rates have persistently remained high (see Siphambe 2003: 481).
However massive, this demographic trend did not depopulate rural areas because Botswana had, at the same time, a high population growth rate. Moreover, and highly significant for central arguments in this volume, the capital is surrounded by five of the seven Tswana royal towns (see below) and a number of large villages, mostly within an hour's drive or less from the capital. Many people employed in urban centres have thus continued to live within the context of family and descent groups or at least kept in close touch with rural family households. The royal towns which have, from precolonial times, been – in an African context – exceptionally large with thousands, if not tens of thousands, of people (Gulbrand-sen 2007), also have growth momentum as district governmental centres, service and trading centres and by some minor industries.
Map 1 displays the administrative division of the Bechuanaland Protectorate as implemented by the British upon colonization in 1885. In the eastern and northwestern part of the country, the divisions are named ‘native reserves’, dominated by the eight Tswana kingdoms which, as already suggested, the British officially recognized and whose rulers were subjected to the colonial administration as instruments of government. The native reserves were, as Map 1 shows, of highly unequal size, territorially and population-wise. The vast areas denoted ‘crown lands’ were extremely sparsely populated by people living scattered in small villages, hamlets and mobile bands. The Tswana-centred native reserves, mainly located in the Eastern part of the Protectorate, were ethnically mixed to a very different extent, with the four largest ones – the Bangwato, the Bakwena, the Bangwakets and the Batawana – comprising vast groups of different origins.
The respective Tswana dikgosi were located in the royal towns of, respectively, Serowe (Bangwato), Molepolole (Bakwena), Kanye (Bangwaketse), Maun (Batawana), Mochudi (Bakgatla), Ramotswa (Malete) and Tlokweng (Batlokwa). The small area in the extreme southeast denoted Barolong farms serving as agricultural lands for the Tswana people of Barolong-Tshidi centred in the royal town of Mafeking on the South African side of the border.
Map 1: Sketch map of Bechuanaland Protectorate
(source: Schapera 1970)
Map 2: Map of the Republic of Botswana
(based on an original map courtesy of the Department of Surveys and Mapping, Government of Botswana, Gaborone)
By comparing Maps 1 and 2 it is readily apparent that there is considerable correspondence between the colonial and postcolonial administrative divisions. This is most evident in the case of Central (Ngwato), Kweneng, Kgatleng, Ngamiland (Tawana) and Southern (which includes the Ngwaketse and the Barolong farms). The Tswana royal towns serve as district administrative centres in all these cases. The Batlokwa and the Bamalete are combined into the small South-East District, with Ramotswa as the district centre. Moreover, the Tswana royal towns host the postcolonial administrative centres – Maun (Ngamiland), Serowe (Central), Molepolole (Kweneng), Kanye (Southern), Mochudi (Kgatleng) and Ramotswa (South-East). The additional, most sparsely populated districts of Kgalagadi, Ghazi, Chobe and North-East do not fall into this pattern.
1. Key contributions include Colclough and McCarthy 1980; Harvey and Lewis 1990; Hillbom 2008; Isaksen 1981; Siphambe et al. 2005; Thunberg-Hartland 1978.
2. See Edge and Lekorwe 1998; Good 1994, 1999a, 2002, 2008; Gulbrandsen 1996a; Gunderson 1970; Holm 1985, 1988; Holm and Molotsi 1989; Maundeni 2002; Mbabazi and Taylor 2005; Molutsi and Holm 1990; Parson 1984; Picard 1985; Samatar 1999; Somolekae and Lekorwe 1998; Sebudubudu 2005; Taylor 2003, 2005; Tsie 1996; Vengroff 1977.
3. Kenneth Good has, in a series of articles and a recent book (Good 2008) represented a major, critical voice of Botswana's political development; this has cost him his residence permit and consequently his professorship at the University of Botswana.
4. E.g. Hansen and Stepputat (2001), Das and Poole eds. (2004) and Friedman (2011).
5. There is, of course, a well-established anthropological practice to approach of political systems in ways that take care of the ‘the local’ while also addressing assemblages of power of a larger scale. What is often recognized as the seminal work on political anthropology – African Political Systems (Fortes and Evans-Pritchard eds. 1940) – provides early examples of such an effort, as do a range of subsequent studies, including Leach (1954), Barth (1959), Geertz (1980), Claessen and Skalnik eds. (1978), Valeri (1985), Kapferer (1988), Trouilot (1990) and Hansen (1999).
6. E.g. Bratton and van de Walle (1997: 61ff).
7. See Ergas 1987: 9ff.
8. The other countries in which traditional authorities were given official recognition at their independence include Nigeria and Malawi beside Botswana.
9. I am, to be sure, not suggesting that diamonds have caused calamities everywhere else in Africa; other cases in point where diamonds have no such consequences include Namibia and South Africa.
10. Ambiguities and ambivalence in the relationship between postcolonial state and indigenous authority figures are clearly reflected in von Rouveroy van Nieuwaal's (1999) comparative discussion of what he refers to as the ‘hybrid role of chieftaincy in postcolonial Africa’. They are, moreover, reflected in a number of case studies, including Cameroon (Awasom 2005), Ghana (Lentz 1998; Rathbone 2000), Mozambique (Bertelsen 2003; Buur and Kyed 2006), Nigeria (Vaughan 2006), South Africa (Oomen 2005), Tanzania (e.g. Bienen 1970), Uganda (Karlstrøm 1996), and Zimbabwe (e.g. Ladley 1991). These ambiguities are interestingly illuminated by Western Zambia, where there has been an opposite trend: Van Binsbergen first explained that ‘state and chieftainships are closely interlocking aspects of modern Zambia’ to the extent that ‘the incumbents of positions of the state…, in their effective exercise of popularly supported power, simply cannot do without chiefs’ (1987: 191–92). Later he related that ‘[w]hereas in the first decades of the postcolonial era they [the royal chiefs] effectively expanded
